sign in sign up
<<
Home , Ibutilide, Vernakalant

Vernakalant NDA 22034 Resubmission Preston M. Dunnmon, MD, MBA, FACP, FACC Division of Cardiovascular and Renal Products “FDA Overview of Cardiovascular Safety” Daniel Woronow, MD, FACC Office of Surveillance and Epidemiology Division of Pharmacovigilance “Safety of or Electrical in Patients with or Flutter” NDA 22034 Resubmission FDA Overview of Cardiovascular Safety Cardiovascular and Renal Drugs Advisory Committee Meeting December 10, 2019

Preston M. Dunnmon, MD, MBA, FACP, FACC Medical Officer www.fda.gov Presentation Agenda • Regulatory history • Mechanistic considerations: Vaughan-Williams Class • Clinical Cardiovascular Safety – Re-integration of controlled clinical trials – SPECTRUM observational registry • Pre-infusion checklist – Will this identify subjects at risk? • Safety of alternative therapies – Ibutilide (approved for the rapid conversion of AFib) – Electrical cardioversion (ECV) • Conclusions www.fda.gov 3 Vernakalant Regulatory History – 1 • 2006: Original submission of NDA 22034 – 375 treated subjects – 8 serious adverse events – 1 death • 2008: Approvable letter – Risk appeared excessive with respect to benefit – Another controlled study requested with entry criteria resulting in <1% risk of serious cardiovascular adverse events • 2009: ACT V was initiated – Planned 474 subjects enrolled (2:1) www.fda.gov 4 Vernakalant Regulatory History – 2 Investigational New Drug Application (IND) Placed on Clinical Hold in 2010

• 2010: Pulseless electrical activity and cardiogenic shock at the end of first vernakalant infusion – Subject at low baseline risk (only hypertension; left ventricular hypertrophy)

• Intravenous (IV) vernakalant IND was placed on full clinical hold www.fda.gov 5 Vernakalant Regulatory History – 3 on Clinical Hold Since 2010

• 2014-2016: applicant failed to identify a new dosing strategy that would be effective without causing the negative inotropic effects

• 2019: NDA resubmitted – Interim controlled safety data – SPECTRUM

www.fda.gov 6 What is Vernakalant?

Applicant’s position: “Vernakalant IV, an atrial-selective blocker, has a differing mechanism of action that mitigates some of the main safety concerns of other anti-arrhythmic treatments.” (2016 SPECTRUM Protocol, Synopsis)

“Vernakalant is a multichannel blocker of certain potassium channels, an atypical class III antiarrhythmic.” (2019, risk management module of NDA) “BRINAVESS is an antiarrhythmic drug that acts preferentially in the atria by prolonging atrial refractoriness and slowing impulse conduction in a rate- dependent fashion… Because of its atrial preferential actions, vernakalant does not readily fit in the Vaughan Williams anti-arrhythmic drug classification, which is based on ventricular activity.” (2019, Proposed label) www.fda.gov 7 Vernakalant is a Non-selective Ion – Affects Ventricles As Well As Atria

Ikur IKACh INa-P INa-L IKr Ito (atrial) (atrial) IC50 (µM) 13 14 7.2 20 3.2 10 at ≤ 1 Hz

Sponsor data. IC50 values shown are the lowest for each current. A safety margin <30-fold is likely to have clinical effects (Redfern et al. Cardiovasc Res. 2003 Apr 1;58(1):32-45.) • All channels blocked at therapeutic concentrations • All channels blocked with indistinguishable potency

www.fda.gov 8 Vernakalant Channel Blocking Profile Is Similar to (Known Class IC)

Channels blocked with similar potencies I I I I I IKur IKAch Na-P Na-L Kr to Ca-L (atrial) (atrial) Vernakalantapplicant        Flecainide1,2,3,4,5,6       

1Guo D and Jenkinson s. J Pharmacol & Toxicological Meth. 2019, 99:106575. 2 Kramer J, Obejero-Paz C, Myatt G, et al. Sci Rep. 2013, 3: 2100. 3 Yue L, Feng J, Wang Z and Nattel, S. Cardiovascular Research. 2000, 46:151-161. 4 Yamashita T, Nakajima T, Hazama H and Kurachi Y. J Pharmacol Exp Ther. 1995, 274: 315-312. 5 Inomata N, Ishihara T and Akaike N. Br J Pharmcol. 1991, 104:1007-1011. 6 Wang D, Sato T and Arita M. Cardiovas Res. 1995, 29: 520-525.

www.fda.gov 9 Vaughan-Williams Classification of Blockers • Class I antiarrhythmic drugs – Prolong the QRS and are negative inotropes – Three subclasses defined by dissociation constant from channel:

Sub-classes Kinetics1,2,3 Dissociation constant3 1A () Intermediate 1-10 seconds 1B () Fast < 1 second 1C (flecainide) Slow > 10 seconds

1Frumin H, Kerin N and Rubenfire M. J Clin Pharmacol. 1989, 29: 387. 2Mittchell L. Drugs for . Merck Manuals. 3Lei M, Wu L, Terrar D and Huang C. Circulation. 2018, 138: 1879. www.fda.gov 10 Vernakalant Dissociates Slowly From the Sodium Channel

RSD1205 • Class IC definition: (vernakalant and its diastereomers)

dissociation time Time (s) constant greater than 10 seconds

• Vernakalant’s calculated Peak inward INa (pA) dissociation time constant is 49.4 seconds Source: sponsor www.fda.gov 11 2014 Canine Contractility Study – Assessed as dp/dt

Flecainide

Vernakalant % decrease in dp/dt max dp/dt in % decrease 0 15 30 45 60 75 90 Time (min)

Dog Exposure Therapeutic Cmax IV Vernakalant (n=8) ~1800 ng/ml 4300 ng/ml IV Flecainide (n=4) ~700 ng/ml 200-1000 ng/ml 12 Canine Deaths

• 2012 dog hemodynamic study (N=11) – One of 6 dogs given IV vernakalant after induction of cardiac dysfunction with 3-5 weeks of rapid ventricular pacing died on-study: • Increase in electrocardiogram (ECG) QRS duration • Rapid decreases in blood pressure (BP), heart rate (HR), and cardiac output (CO); dog lost consciousness • Regained consciousness, but HR and BP fell precipitously again; dog could not be recovered • 2014 dog contractility (dp/dt) study (N=8) – The only dog that received IV vernakalant after one week of rapid atrial pacing was found dead in its cage within 2 hours of vernakalant administration.

13 Vernakalant Prolongs the QRS Interval in Humans Cumulative percentage Cumulative 0 25 50 75 100 -50 0 50 100 Maximum change in QRS interval (msec) Vernakalant arm stratified by # of doses and outcome of conversion 14 Summary – Mechanistic Considerations • Vernakalant is: – A Vaughan-Williams Class IC agent – A negative inotrope that prolongs the QRS – Not atrial specific (affects ventricles and atria) – Similar to flecainide

• Vernakalant would be expected to cause serious cardiovascular events such as: – Hypotension – – Ventricular arrhythmias – Possible death www.fda.gov 15 Vernakalant Safety: Controlled Clinical Trials

www.fda.gov 16 IV Vernakalant-related serious adverse events were consistent with the safety profile of a Vaughan-Williams Class IC antiarrhythmic drug Serious adverse event of Placebo Vernakalant interest – 0-2 hours post-dose (N=459) (N =1073) Hypotension 0 9 (0.8%) 0 16 (1.5%) 0 3 (0.3%) Bradycardia 0 6 (0.6%) Ventricular arrhythmia 0 5 (0.5%) Conduction disturbance 0 3 (0.3%) Vernakalant-related death 0 2 (0.2%) www.fda.gov Data source: all patients population from the integrated safety data for phase 2/3 clinical studies 17 A subset of patients were identified with a greater risk for vernakalant-related adverse effects but could not be prospectively identified

• Subjects who did not convert to sinus rhythm and who did not receive the second dose (N=43, 4%) had worse outcomes as follows: – Higher incidence (26%) of SAEs within 2 hours post-dose – Marked increase in QTcF (>30 msec) and QRS intervals (~20 msec)a – 32% had systolic BP decrease ≥ 20% – 16% had systolic BP decrease ≥ 20% and to < 90 mmHg

• No demographic or disease-specific characteristics were found to prospectively identify most of these subjects www.fda.gov a Placebo-adjusted increase at peak 18 Summary – Controlled Safety - 1

• Vernakalant prolongs the QRS interval in clinical trials • Vernakalant causes adverse effects consistent with its Vaughan-Williams Class IC sodium channel blockade • Most patients who will do poorly with vernakalant cannot be prospectively identified – Harm caused cannot be reliably predicted

www.fda.gov 19 Summary – Controlled Clinical Trial Safety - 2

• Reliable risk mitigation for serious cardiovascular adverse events could not be achieved on the basis of demographic characteristics – Harm cannot be prevented through risk mitigation

• In ACT V, serious hypotension without bradycardia occurred that was unresponsive to pressors for 40 minutes – When harm occurs – in some cases, it cannot be treated

www.fda.gov 20 Vernakalant Safety: SPECTRUM

www.fda.gov 21 Overview of SPECTRUM

• Observational registry for patients who received vernakalant IV in 6 western European countries – N = 1778 – 2,009 vernakalant treatment episodes – 79% prospective and 21% retrospective patients • Data were largely collected through medical chart abstraction

www.fda.gov 22 Incidence of Serious Adverse Events of Special Interest Within 2 hours Post-dose: SPECTRUM versus Controlled Clinical Trials

Serious adverse event SPECTRUM Clinical Trials grouping (N=2,009) (N = 1,073) Hypotension 4 (0.2%) 9 (0.8%) Arrhythmia 18 (0.9%) 16 (1.5%) Bradycardia 7 (0.3%) 6 (0.6%) Ventricular arrhythmia 2 (0.1%) 5 (0.5%) Conduction disturbance 4 (0.2%) 3 (0.3%) Vernakalant-related deaths 0 (0.0%) 2 (0.2%) www.fda.gov 23 SPECTRUM results are not reassuring • SPECTRUM limitations: – Potential selection bias (enrollment based on physician discretion) – Unknown whether all subjects who may have been eligible for vernakalant treatment were actually screened for enrollment – Non-consecutive enrollment (21% of screened subjects not enrolled) – Retrospective enrollment (21%)

• SPECTRUM safety results were consistent with clinical studies: – Incidence of serious adverse events of interest lower than observed in clinical studies, but… – Underreporting of adverse events could not be ruled out in SPECTRUM

www.fda.gov 24 Vernakalant Safety: Pre-infusion Checklist

www.fda.gov 25 Problematic Pre-infusion Checklist Questions/Statements • Labeled Contraindications: – Low baseline blood pressure? – Low baseline heart rate? – Long QT interval? – or known moderate to severe left ventricular dysfunction? • Has the patient received an intravenous rhythm control antiarrhythmic drug (class I and class III) within 4 hours prior to or within 4 hours following, BRINAVESS administration?

• Use of IV BRINAVESS with beta-blockers is not recommended within 2 hours prior to, or 2 hours after, administration. 26 The proposed pre-infusion checklist will not reliably predict which subjects will experience cardiovascular serious adverse events with vernakalant.

www.fda.gov 27 Safety of Ibutilide or Electrical Cardioversion in Patients with Atrial Fibrillation or Flutter Cardiovascular and Renal Drugs Advisory Committee Meeting December 10, 2019 Daniel Woronow, MD, FACC Medical Officer Division of Pharmacovigilance Office of Surveillance and Epidemiology www.fda.gov Purpose • Review available evidence from medical literature and postmarket case reports to determine if there is a substantial risk of death or severe hypotension with – Ibutilide Pharmacological Cardioversion (PCV) or – Electrical Cardioversion (ECV)

www.fda.gov 29 NO CONCLUSIVE EVIDENCE that ECV or Ibutilide PCV causes • Non-embolic fatalities, or • Severe hypotension in patients meeting the ACT V study enrollment criteria such as absence of • History of heart failure (HF) • Significant valvular stenosis • within the preceding 30 days • Clinically significant illness www.fda.gov 30 ACT V Study

• Initiated by the Sponsor to address FDA’s concerns regarding the safety of intravenous vernakalant with respect to serious drug-induced hypotension, bradycardia, and arrhythmias • Primary objective was to evaluate the safety of vernakalant injection in subjects with AFib and no evidence or history of HF • INCLUSION/EXCLUSION CRITERIA FOR ACT V STUDY WERE MORE RESTRICTIVE THAN IBUTILIDE OR ECV STUDIES

www.fda.gov 31 ECV Preferred over PCV • American College of Cardiology/American Heart Association Guidelines (2014): – ECV is preferred [over PCV] in patients with  decompensated HF These are ACT V Exclusion Criteria and  ongoing myocardial ischemia proposed Vernakalant  hypotension Contraindications – There are NO patient subgroups for whom a PCV strategy is preferred over ECV • ECV used more commonly than PCV among surveyed US cardiologists, emergency physicians, and hospitalists regarding acute management of AFib (Funk, 2015) www.fda.gov 32 Ibutilide Premarketing Phase II/III Studies Treatment-Emergent Medical Events With Frequency of More Than 1% and Higher Than That of Placebo Absolute Risk Placebo N=127 All Ibutilide N=586 Difference Event Patients Patients % n % n % CARDIOVASCULAR Sustained polymorphic VT — — 10 1.7 1.7 Non-sustained monomorphic VT 1 0.8 29 4.9 4.1 Non-sustained polymorphic VT — — 16 2.7 2.7 QT segment prolonged — — 7 1.2 1.2 Bradycardia 1 0.8 7 1.2 0.4 Bundle branch block — — 11 1.9 1.9 Atrioventricular block 1 0.8 9 1.5 0.7 Hypotension 2 1.6 12 2.0 0.4 IMPORTANTLY, there were NO deaths in these studies. Instances of sustained polymorphic Ventricular (VT) were all treated successfully. 33 Ibutilide Postmarketing Randomized Controlled Trials (RCTs)

Volgman 1998 Reisinger 2004 Zhang 2005 TOTAL Ibutilide US (Rush- Country Presbyterian-St. Austria China - Luke’s) Ibutilide treatment N=60 N=106 N=41 N=207 arm Ventricular arrhythmia 1 (1.7%) 0 0 1 (0.5%) Polymorphic VT requiring intervention treated successfully

Hypotension 0 0 0 0

Death 0 0 0 0 www.fda.gov 34 Ibutilide Postmarket Case Reports Assessment of Fatal Outcomes • FDA Adverse Event Reporting System (FAERS) database searched – Ibutilide and Outcome of Death – Since US market approval in 1995 to September 2019 • 14 reports (after excluding 2 reports because of insufficient information to determine a causal association) • The 14 reports were heavily confounded and included patients with – “Do Not Resuscitate” orders – patients meeting ACT V exclusion criteria www.fda.gov 35 ECV Safety in RCTs Comparing ECV to PCV

Bellone 2012 Chen 2013 de Paola 2003 Mattioli 1998 Total, ECV

Country Italy China Brazil Italy -

ECV treatment arm N=121 N=59 N=67 N=34 N=281

Ventricular arrhythmia requiring none reported none reported none reported none reported none reported intervention

Hypotension none reported none reported none reported none reported none reported

none reported none reported Mechanical (1 hypoxia, no (1 hypoxia, no respiratory assistance none reported none reported none reported mechanical assistance mechanical assistance or pulmonary edema reported) reported)

Death 0 0 0 0 0 www.fda.gov 36 ECV Observational Studies that include Respiratory or Pulmonary Edema Adverse Events Botkin 2003 Burton 2004 Davarashvili 2018 Country US, Loyola, Chicago US, multihospital Israel Study method retrospective observational, retrospective observational, retrospective observational, outpatient ECV, AFib & atrial consecutive ED-ECV patients, consecutive ECV patients, AFib flutter AFib ECV procedures N=532 N=388 N=1696 HF related baseline 44% average left ventricular INCLUDED unstable within total patient population: characteristics ejection fraction (LVEF) and hypotensive patients • 42% with , moderate or severe • 13% with LVEF < 40% Ventricular arrhythmia none reported 2 (0.5%) none reported requiring intervention Hypotension none reported 0 none reported Mechanical respiratory 2 (0.4%) intubations no intubations No intubations reported assistance or pulmonary 66 (3.9%) developed pulmonary edema edema Death 0 0 no ECV deaths reported (9 deaths of unreported causes within 30 days of ECV) www.fda.gov 37 No Conclusive Evidence of Non-Embolic Fatalities with ECV 33,177 total ECV procedures • 58 total publications containing ECV survival information • 2 “non-sudden cardiac death[s]” reported in Euro Heart Survey Registry study among 712 ECV procedures - Insufficient information to determine a causal association • NO conclusive evidence that Death is a substantial risk with ECV among patients treated for the rapid conversion of atrial arrhythmias to sinus rhythm www.fda.gov 38 Summary Ibutilide PCV Literature review did not identify any instances of ibutilide-related death during index hospital inpatient/outpatient stays among patients who otherwise could have been enrolled in ACT V ECV ECV is generally successful in rapidly converting AFib to sinus rhythm

ECV literature review did not identify any deaths causally related to ECV despite most of these studies including patients with more severe baseline comorbidities than in the ACT V study ECV related serious adverse events (AEs) that are non-transient, and not self- limited occur uncommonly or rarely, despite most of these ECV studies including patients with more severe baseline comorbidities than in the ACT V study www.fda.gov 39 References, page 1

• Bellone A, Etteri M, Vettorello M, et al. Cardioversion of acute atrial fibrillation in the emergency department: a prospective randomised trial. Emerg Med J 2012;29(3):188-91. • Botkin SB, Dhanekula LS, Olshansky B. Outpatient cardioversion of atrial arrhythmias: efficacy, safety, and costs. Am Heart J 2003;145(2):233-8. • Burton JH, Vinson DR, Drummond K, et al. Electrical cardioversion of emergency department patients with atrial fibrillation. Ann Emerg Med 2004;44(1):20-30. • Chen WS, Gao BR, Chen WQ, et al. Comparison of pharmacological and electrical cardioversion in permanent atrial fibrillation after prosthetic cardiac valve replacement: a prospective randomized trial. J Int Med Res 2013;41(4):1067- 73. • Corvert Label. Inc. New York, NY. Revised February 2017. • Davarashvili I, Acha MR, Glikson M, et al. Pulmonary Congestion Complicating Atrial Fibrillation Cardioversion. Am J Cardiol 2018;122(10):1701-06. • de Paola AA, Figueiredo E, Sesso R, et al. Effectiveness and costs of chemical versus electrical cardioversion of atrial fibrillation. Int J Cardiol 2003;88(2-3):157-66. • El-Am EA, Dispenzieri A, Melduni RM, et al. Direct Current Cardioversion of Atrial Arrhythmias in Adults With Cardiac Amyloidosis. J Am Coll Cardiol 2019;73(5):589-97. • Funk AM, Kocher KE, Rohde JM, et al. Variation in practice patterns among specialties in the acute management of atrial fibrillation. BMC Cardiovasc Disord 2015;15:21. • Gallagher MM, Yap YG, Padula M, et al. Arrhythmic complications of electrical cardioversion: relationship to shock energy. Int J Cardiol 2008;123(3):307-12. www.fda.gov 40 References, page 2

• Grönberg T, Nuotio I, Nikkinen M, et al. Arrhythmic complications after electrical cardioversion of acute atrial fibrillation: the FinCV study. Europace 2013;15(10):1432-5. • Guédon-Moreau L, Gayet JL, Galinier M, et al. Incidence of early adverse events surrounding direct current cardioversion of persistent atrial fibrillation. A cohort study of practices. Therapie 2007;62(1):45-8. • Hellman T, Kiviniemi T, Vasankari T, et al. Prediction of ineffective elective cardioversion of atrial fibrillation: a retrospective multi-center patient cohort study. BMC Cardiovasc Disord 2017;17(1):33. • January CT, Wann LS, Alpert JS, et al. 2014 AHA/ACC/HRS guideline for the management of patients with atrial fibrillation: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and the Heart Rhythm Society. J Am Coll Cardiol 2014;64(21):e1-76. • Mattioli AV, Castelli A, Andria A, et al. Clinical and echocardiographic features influencing recovery of atrial function after cardioversion of atrial fibrillation. Am J Cardiol 1998;82(11):1368-71. • Pisters R, Nieuwlaat R, Prins MH, et al. Clinical correlates of immediate success and outcome at 1-year follow-up of real-world cardioversion of atrial fibrillation: the Euro Heart Survey. Europace 2012;14(5):666-74. • Reisinger J, Gatterer E, Lang W, et al. Flecainide versus ibutilide for immediate cardioversion of atrial fibrillation of recent onset. Eur Heart J 2004;25(15):1318-24. • Steinberg BA, Schulte PJ, Hofmann P, et al. Outcomes after nonemergent electrical cardioversion for atrial arrhythmias. Am J Cardiol 2015;115(10):1407-14. • Volgman AS, Carberry PA, Stambler B, et al. Conversion efficacy and safety of intravenous ibutilide compared with intravenous procainamide in patients with atrial flutter or fibrillation. J Am Coll Cardiol 1998;31(6):1414-9. • Zhang N, Guo JH, Zhang H, et al. Comparison of intravenous ibutilide vs. for rapid termination of recent onset atrial fibrillation. Int J Clin Pract 2005;59(12):1395-400. www.fda.gov 41 Vernakalant NDA 22034 Resubmission

FDA Conclusions

Preston M. Dunnmon, MD, MBA, FACP, FACC FDA Conclusions - 1

Vernakalant • Is a Vaughan-Williams Class IC antiarrhythmic • Is not atrial-specific (affects ventricles and atria) • Prolongs the QRS, markedly so in some subjects • Is a negative inotrope in dogs and in humans • Has caused deaths in dogs and in humans

43 FDA Conclusions - 2 Vernakalant • Is similar to flecainide – in dogs, vernakalant’s negative inotropic effect is as large as that observed with IV flecainide but does not recover during 90 minutes of post-dosing observation – in humans, adverse events are similar (hypotension, bradycardia, ventricular arrhythmias, atrial flutter, conduction disturbances, and death)

44 FDA Conclusions - 3

• The proposed pre-infusion checklist will not reliably predict which subjects will experience cardiovascular SAEs with vernakalant

• SPECTRUM results are not reassuring regarding vernakalant’s cardiovascular safety

45 FDA Conclusions - 4 • Vernakalant has induced harm that cannot be reliably: – Predicted – Prevented – Treated (in some cases) • In contrast to vernakalant, ECV and ibutilide PCV can cause adverse events that are transient or treatable • We believe the benefit risk profile of vernakalant is unfavorable for the proposed indication

46

BACK-UP SLIDES SHOWN

48 Patient “A” Clinical Course

• Echo/ECG documented PEA – 40 minute pulseless resuscitation Echo Day LVEF RVEF MR • Encephalopathy (never regained consciousness) D1 (AFib 156 bpm) 44% WNL Mild • Renal failure requiring dialysis D1 (CPR) 0 0 0 • Recurrent AFib during dialysis D1 (SR post ECV) 25% NR NR • Hepatic failure • Rhabdomyolysis D2 (AFib 96bpm) 40% WNL Mild • Coagulopathy D19 (SR 88 bpm) 49% Mild • Aspiration D 23 78% WNL Severe • Sepsis • Death - hypovolemic shock Day 28: Death (ischemic/necrotizing colitis and resultant uncontrollable GI bleeding) 50 Divergent Assessments of Patient A Applicant: FDA: “The follow-up echocardiogram showing left ventricular hypertrophy, a This subject’s rapid recover of left left atrial size of 50 mm and mitral ventricular function following his regurgitation suggests long standing vernakalant-induced, 40-minute, mitral regurgitation secondary to left pulseless arrest does not support the ventricular dysfunction and an ejection applicant’s explicit assessment that fraction of 25%. With normal coronary alcoholic cardiomyopathy was to anatomy, the most likely etiology is blame, or the implicit idea that an cardiomyopathy and long- alcoholic cardiomyopathy could have standing hypertension. It is very been identified by either ACT V unusual to develop either renal failure, exclusion criteria or the proposed liver failure, or rhabdomyolysis or pre-infusion checklist that could coagulopathy after cardiac have or should have kept this person resuscitation suggesting that he had a from being dosed with vernakalant. very prolonged state of inadequate circulation.” 51 Postmarket Serious Hypotension

• Patient “B” with no history of cardiac disease presented with new and recent-onset AFib. He received Brinavess 318 mg IV (one dose only) and experienced the following at times post-infusion: – 5 min: itching, clammy, diaphoresis – 10 min: SBP fell 130 to 80 mmHg while in AFib with HR 100-130 – 12 min: 500 cc NS bolus – 15-25 min: sweating, clammy, SBP 70-80, HR 100 (AFib). Second 500 cc NS bolus – 25 min: deterioration of hemodynamic status, very clammy, tonic-clonic seizures, loss of consciousness, no carotid pulse, arrest called, and chest compressions initiated – 27 min sinus brady @ 40-50/min, narrow QRS, SBP @ 80/min, consciousness recovered

• Echocardiograms (all by same experienced “highly competent” sonographer) showed: – Echo 1: 30 minutes post drug infusion LVEF “low normal” – Echo 2: 90 minutes post drug patient now awake in “good stable condition clinically,” LVEF < 20% – Echo 3: 300 minutes (5 hours) post drug, normal echo with normal LV systolic function, troponins negative

52 Additional ECV Deaths Unlikely or Unclear Causal Association 32 additional deaths: could not determine to be related to the ECV procedure • (Guédon-Moreau, 2007) 3 (0.4%) deaths …………………………………..…………………… among 684 ECV procedures – Lethal brain hemorrhage in a patient on both LMWH and warfarin – 86-year-old patient with hypertrophic cardiomyopathy who died of HF one day after ECV – 78-year-old patient with valvular cardiomyopathy who died one month after ECV • (Hellman, 2018) 4 (0.1%) deaths of unreported causes within 30 days …....……. among 4356 ECV procedures • (Steinberg, 2015) 14 (1.4%) deaths within 30 days…...... … among 1017 ECV procedures – 5 from HF – 2 from respiratory failure – 2 from septic shock – 5 cause not available • (Gallagher, 2008) 9 (0.3%) unexpected deaths within 28 days after attempted ECV….… 2522 ECV procedures – 2 attributed to pulmonary embolism. These patients were older than the population mean. • (Grönberg, 2010) 1 (0.01%) death due to aortic dissection 24 hours after ECV... among 6906 ECV procedures • (El-Am, 2019) 1 (0.7%) death, patient with amyloidosis who developed left hemiplegia the same night following successful ECV and died 5 days later ……………………………………………….. among 148 ECV procedures www.fda.gov Above studies all Observational 73 Sensitivity Analysis

• The 6 studies from the previous slide are included in the 33,177 total ECV procedures previously mentioned • Irrespective of causal association or patient comorbidities, there are 43 total deaths combining studies by Pisters, Davarashvili, Guédon- Moreau, Hellman, Steinberg, Gallagher, Grönberg, and El-Am • This calculates to 43/33,177=0.13%, which represents all deaths, maximum follow-up of 30 days, irrespective of causal relationship, and irrespective of patient comorbidities • There is no conclusive evidence that any of these deaths are causally related to ECV www.fda.gov 74 ECV Publications Compared with Vernakalant Clinical Trials Although we found NO deaths conclusively caused by ECV, if we extend our analysis to ECV deaths IRRESPECTIVE of causal association, there remains a very low number of deaths with ECV compared with Vernakalant clinical trials. ECV study follow-up times are up to 30 days.

Vernakalant Clinical Trials 58 ECV publications

AE Placebo Vernakalant ECV Grouping (N=459) (N=1073) (N=33,177) Deaths 1 (0.2%) 8 (0.7%) 43 (0.13%) www.fda.gov 75